[Ffmpeg-devel-irc] ffmpeg.log.20111202

[burek]

[00:52] <shysoft> Is there a trick to using debugging symbols with ffmpeg?
[00:52] <shysoft> I'm having trouble stepping into the libav* libraries.
[01:03] <teratorn> shysoft: yeah there is kind of a trick
[01:04] <teratorn> shysoft: this is what I use: --enable-debug --disable-stripping --disable-asm --disable-optimizations
[01:04] <shysoft> teratorn: thanks. I'll try that.
[01:42] <pasteeater> relaxed: you're right. trampst...i mean timestamp has been depreciated for creation_time. where the hell have i been?
[02:20] <abstrakt> how do I know if this avi is 1080i or 1080p
[02:20] <abstrakt> here's the output of ffmpeg -i
[02:20] <abstrakt> Stream #0.0: Video: h264 (Main), yuv420p, 1920x1080 [PAR 1:1 DAR 16:9]
[02:51] <SMJ> http://paste.servut.us/plain/5v0z can someone explain what operation is not permitted?
[02:55] <SMJ> ah... just when I thought Ubuntu had a decently recent version of ffmpeg, it's not even at the same major version...
[03:36] <JackD> Good Night/Day/Morning
[03:37] <JackD> How can I get video from my camera and create a jpg every second in a folder?
[03:38] <JackD> I had tried ffmpeg -f vfwcap -s 640x480 -r 1 -i 0 output%d.jpg but it make the jpg change every second not be created every second
[03:38] <JackD> In true it create a big jpeg sequence equal the framerate of my camera.
[03:44] <kanzure> jackd: %03d maybe?
[05:24] <Whitewizard> Moar problems! when i seek to a certain point in a video the audio becomes desynced! using -ss 00:10:42
[05:34] <rynop> im using ffmpeg to convert a glc-capture format file to libx264. Every preset i use seems to create a video that "dark" - like not as bright as the original. Total ffmpeg noob here, is this a common problem?
[05:36] <rynop> http://pastebin.com/82XgZ0rc is an example of what im doing that is producing a dark video.
[05:46] <pasteeater> Whitewizard: option placement is important. did you use it as an input or output option?
[05:46] <Whitewizard> before the -i
[05:46] <pasteeater> use a pastebin site to show your ffmpeg command and the complete console output
[05:47] <pasteeater> rynop: same for you
[05:47] <Whitewizard> like its weird.. i got 5 seconds of video, that i want, but an extra 5 seconds of audio before it
[05:47] <rynop> pasteeater, i did post my pastebin..
[05:47] <hi117> it still had pipes included
[05:48] <rynop> pasteeater, or u want full output
[05:48] <Whitewizard> http://pastebin.com/gub5LcZ1
[05:49] <pasteeater> rynop: yes, include command and complete output
[05:50] <pasteeater> Whitewizard: where's your command?
[05:50] <Whitewizard> lets try this again..
[05:51] <Whitewizard> http://pastebin.com/7RcnQj9W
[05:51] <rynop> ah sry. http://pastebin.com/B9QwnVnC and also tried http://pastebin.com/Y2QSyBdL . Same "dark" results.
[05:51] <Whitewizard> yes, i spelt converter wrong, but its too late to change
[06:00] <pasteeater> rynop: does it look normal with ffplay? ffplay video.mp4
[06:03] <rynop> pasteeater, nope still darker than the original (glc-play 1092-3.glc). This is a recording of a video game (starcraft 2), its a pretty "dark" game to begin with, not sure if that matters when the encoder is compressing.
[06:03] <Whitewizard> the effect seems proportionate  to how many frames were dropped..
[06:03] <pasteeater> i'm not familiar with glc-play. there might be some colorspace conversion going on.
[06:04] <rynop> pasteeater, is there some sort of "dark_mask" option in ffmpeg? reading http://www.mplayerhq.hu/DOCS/tech/encoding-tips.txt seems like my problem is simliar to this menucoder "dark_mask"
[06:04] <pasteeater> Whitewizard: did you try moving -ss after -i input?
[06:06] <Whitewizard> it jsut forever encodes at 0fps
[06:07] <pasteeater> wait for it. that method is slower, but it will help rule out the cause.
[06:08] <pasteeater> it appears to be sitting there, but it's acrually decoding until it reaches your -ss value.
[06:08] <pasteeater> rynop: i'm not sure
[06:08] <rynop> pasteeater, I did just find a -lumi_mask option in ffmpeg, messign w/ that
[06:09] <pasteeater> you could try: ffmpeg -i - -vframes 1 output.png
[06:10] <pasteeater> to test if rgb looks normal.
[06:10] <rynop> frames X -  where X is the number of the frame in the video? higher means later in vid?
[06:11] <pasteeater> -vframes = number of frames to output
[06:11] <pasteeater> 1 for 1 image.
[06:11] <rynop> ah ok. early in the vid (the loading screen) the brightness is fine. its just in game. So i realy need to get some frames from the middle of my vid to compare
[06:12] <pasteeater> you should also try a more recent ffmpeg: http://ubuntuforums.org/showthread.php?t=786095
[06:12] <rynop> pasteeater, ok good idea.
[06:15] <Whitewizard> that seemed to haveworked
[06:15] <Whitewizard> odd..
[06:15] <Whitewizard> thnaks
[06:15] <Whitewizard> thanks
[06:16] <pasteeater> you might be able to speed up decoding with -threads 4 or some appropriate number as an input option.
[06:17] <pasteeater> rynop: you could add -ss 23 if you want ffmpeg to skip the first 23 seconds so you have time to get from menu to game
[06:19] <pasteeater> (as an output option)
[06:25] <rynop> pasteeater, -lumi_mask -dark_mask seem to be helping. Just messing around with values to get teh right setting, not documented well
[06:26] <rynop> so i lied its not helping, was comparing different frames.
[06:32] <pasteeater> did you try the png test? did you try a newer ffmpeg?
[06:34] <rynop> no to both. not yet. didnt feel like going thru teh ffmpeg compile
[06:35] <rynop> watned to try some of these other options before getting into compiling. looked for a launchpad ffmpeg repo - found some but they are way back level, not sure whats up with that.
[07:48] <alyawn> I'm attempting to write a simple filter and every frame gets dropped. Is there a way to find out why ffmpeg is dropping my frames?
[08:22] <relaxed> alyawn: I think the select filter will do that.
[08:24] <relaxed> oh, nm
[08:25] <relaxed> alyawn: loglevel may help
[08:52] <cryptopsy> yesterday i couldn't convert vob to mp3 because a libmp3lame was missing but i'm able to convert avi to mp3 no problems
[09:03] <relaxed> cryptopsy: I recall.
[09:03] <cryptopsy> yes, so what's libmp3lame?
[09:03] <cryptopsy> why do i need it?
[09:03] <cryptopsy> exactly
[09:04] <cryptopsy> relaxed: actually in some circumstances i can produce an output file (mp3) 
[09:04] <cryptopsy> that is non-0 size
[09:04] <cryptopsy> for example my vob is 1giga, the output file is a 220mega mp3
[09:04] <relaxed> are you copying?
[09:04] <cryptopsy> -acodec copy , if that's what you mean
[09:05] <cryptopsy> i just want to extract the audio 
[09:06] <relaxed> ffmpeg -i input.vob output.flac
[09:08] <cryptopsy> relaxed: 0 size
[09:08] <cryptopsy> -rw-r--r-- 1 root root 1.0G Dec  1 00:56 vts_01_1.vob
[09:09] <cryptopsy> -rw-r--r-- 1 root root    0 Dec  2 04:07 test.flac
[09:10] <relaxed> can you pastebin the command and output?
[09:10] <cryptopsy> can i just paste to the channel?
[09:10] <cryptopsy> i'll start pastebining when others are talking
[09:11] <relaxed> no
[09:11] <cryptopsy> k what pastebin do i use?
[09:11] <cryptopsy> oh it's back up
[09:11] <cryptopsy> http://pastebin.com/VHi08duX
[09:12] <relaxed> Did you compile ffmpeg?
[09:13] <cryptopsy> sure
[09:13] <relaxed> why is everything disabled?
[09:14] <cryptopsy> 0.8.6(04:02:24 12/02/11)(mp3 -3dnow -3dnowext -X -aac -aacplus -alsa -altivec -amr -avx -bindist -bzip2 -celt -cpudetection -debug -dirac -doc -encode -faac -frei0r -gsm -hardcoded-tables -ieee1394 -jack -jpeg2k -mmx -mmxext -network -oss -pic -qt-faststart -rtmp -schroedinger -sdl -speex -ssse3 -static-libs -test -theora -threads -truetype -v4l -vaapi -vdpau -vorbis -vpx -x264 -xvid -zlib)
[09:15] <relaxed> I'm 12 and what is this?
[09:15] <cryptopsy> the compile options for ffmpeg
[09:15] <cryptopsy> so it is compiled with mp3 support, but not 3dnow, etc
[09:16] <cryptopsy>  + + mp3              : Add support for reading mp3 files
[09:16] <relaxed> OH! this is gentoo?
[09:16] <cryptopsy> yes
[09:16] <cryptopsy> sure
[09:17] <relaxed> good because the way you're building it is completely broken
[09:17] <cryptopsy> ah okay that would explain it
[09:17] <cryptopsy> i updated the repo and ffmpeg upgraded to a new version
[09:17] <cryptopsy> everything is fucked and busted in the tree nowadays, i don't know how it happened
[09:17] <cryptopsy> how were you able to tell?
[09:18] <relaxed> Because on pastebin I could see all the --disable-* in your output. Plus you can't seem to encode simple audio.
[09:19] <cryptopsy> unfortunatley i don't have the .avi i keep talking about
[09:22] <relaxed> lol, wtf is "-encode"? Do you have encoding disabled?
[09:23] <cryptopsy>  - - encode           : Adds support for encoding of audio or video files
[09:23] <cryptopsy> new flag
[09:23] <cryptopsy> it was builtin in the old version
[09:23] <cryptopsy> gentoo doesn't warn users about these changes
[09:23] <cryptopsy> thanks
[09:24] <relaxed> most of those use flags are wrong.
[09:24] <cryptopsy> i'll show you the list
[09:25] <relaxed> pastebin
[09:25] <cryptopsy> i'm getting sick of pastebin, now it's requiring people to break captchas
[09:27] <cryptopsy> http://pastebin.com/zFSt6SnW
[09:29] <cryptopsy> ompldr.org/vYmpzaA
[09:30] <cryptopsy> the second link works, lists all the use flags and their descriptions
[09:31] <relaxed> add support for all of these http://pastebin.com/WXMzEc4h
[09:32] <cryptopsy> we could go to mp3
[09:32] <cryptopsy> pm*
[09:33] <cryptopsy> pastebin are not loading for me
[09:34] <cryptopsy> but with encode seems to work
[09:34] <cryptopsy> slowly
[09:34] <cryptopsy> it was much faster for avi
[10:01] <Guest26588> hi there anyone alive?
[10:06] <giany> hum?
[10:06] <giany> any idea what this means : "[mov,mp4,m4a,3gp,3g2,mj2 @ 0x9e71340] moov atom not found"
[10:06] <giany> when checking a converted movie?
[13:58] <vega13> what does a videoparser do exactly?
[14:15] <vega13> never mind
[14:22] <zmitya> hi Gents
[14:22] <zmitya> yesterday I have posted this: https://lists.ffmpeg.org/pipermail/ffmpeg-user/2011-December/003445.html
[14:22] <zmitya> do you guys have any comments on this ? or did I post to a wrong place ?
[14:22] <aacffmpeg> hi, question: i have an ios application that grabs audio and converts it to AAC. What I want is to feed FFMPEG with these AAC packages and send them via ffmpeg to my MPEG-TS stream. How can I go about this? Any references?
[15:18] <id10t> 'lo all
[15:20] <id10t> anyone know if there is a way to save a movie from the visualizations that ffplay produces for a mp3 file?  IOW, play mp3 file, dump visualizations to a movie file w/ mp3 file as audio track
[15:36] <simon3z> hi all I have a mpeg-ts stream (from dvb-t) that has some damaged parts and some non-monotonic dts... is there any way to let ffmpeg fix this problems copying the streams to a mpeg-ps file? (using "-vcodec copy -acodec copy -f vob" the output has the a/v streams not synchronized)
[16:58] <cryptopsy> i convert .vob to mp3 with ffmpeg -i my.vob -ac 2 -ab 320k out.mp3 , and i can see it encode with bitrate=320k, but 'file out.mp3' shows
[16:58] <cryptopsy> Audio file with ID3 version 2.4.0, contains: MPEG ADTS, layer III, v1,  56 kbps, 48 kHz, Stereo
[16:58] <cryptopsy> is that a bitrate of 56kbps i see there?
[17:07] <teratorn> I have a C/ffmpeg program that does a basic transcoding job... but I'm having trouble with the resulting file. input and output are both webm. ffprobe shows "1000k" fps for the input, and "1kfps" for the output file - and the output file plays as fast as the cpu can render it... any clue what could be wrong?
[18:14] <beandog> hey guys which ML would be the best one to send a feature request to
[18:16] <cryptopsy> ?
[18:16] <beandog> !
[18:17] <beandog> specifically, wanted to post the idea of getting qt-faststart in ffmpeg, along with an option to fix the metadata automatically at end of encoding with an -http option or something to make it stream-happy
[18:34] <twid> when i try to convert file using ffmpeg i get "mjpeg: unsupported coding type (cf)"... what does it mean by that?  thx....
[18:55] <cryptopsy> how can i confirm the bitrate of an mp3 ffmpeg produced?
[18:58] <twid> cryptopsy : You can use ffprobe to check bitrate of stream....
[19:00] <simon3z> hi all I have a mpeg-ts stream (from dvb-t) that has some damaged parts and some non-monotonic dts... is there any way to let ffmpeg fix this problems copying the streams to a mpeg-ps file? (using "-vcodec copy -acodec copy -f vob" the output has the a/v streams not synchronized)
[19:03] <cryptopsy> twid: thanks
[19:04] <twid> NO prb..
[19:06] <cryptopsy> it's telling me 320k bit rate but file seems to say 56k
[19:06] <cryptopsy> file 2.mp3 
[19:06] <cryptopsy> 2.mp3: Audio file with ID3 version 2.4.0, contains: MPEG ADTS, layer III, v1,  56 kbps, 48 kHz, Stereo
[19:06] <cryptopsy> indeed ffmpeg is reporting correctly "bitrate: 320 kb/s"
[19:06] <cryptopsy> what is that file is telling me?
[19:20] <JackD> Hello
[19:23] <JackD> Can someone help me with a simple task with ffmpeg cli?
[19:24] <pasteeater> beandog: you can add a feature request to the bug tracker
[19:24] <pasteeater> https://ffmpeg.org/trac/ffmpeg/
[19:24] <beandog> pasteeater: kk thanks
[19:25] <pasteeater> type: enhancement, priority: wish
[19:28] <JackD> weird nicks ...
[19:32] <JackD> I think I mdo not need to past it on pastebin - is is small - but if I really need... Well I want to captire a frame every second and put the image in a jpg file. I tried "ffmpeg -f vfwcap -s 640x480 -r 1 -i 0 output%d.jpg" but it only refresh each second it do not create the jpg each second.
[19:34] <JackD> How can I do it, or where I can find info about it? I read some docs but I do not went so far.
[19:35] <beandog> pasteeater: okay
[19:39] <teratorn> anyone have a clue why ffmpeg might produce a file which it is then unable to process? http://codepad.org/ijvbPYMc
[19:42] <JackD> teratorn: have an error about your framerate in the output, have you figured it?
[19:44] <teratorn> JackD: I don't know. I think it's a bug that ffmpeg produces a file that it chokes on
[19:50] <pasteeater> teratorn: i can duplicate the issue if I add -r 0.08
[19:51] <pasteeater> also, sameq does not mean same quality, and you shouldn't be using it in this case as vp8 and h264 probably don't use the same quantizer scale
[19:52] <pasteeater> ffmpeg -h | grep sameq: -sameq    use same quantizer as source (implies VBR)
[19:52] <teratorn> I see
[19:53] <pasteeater> so either use -b:v or -crf when encoding from h264 to vp8
[19:54] <teratorn> OK - well I still have the same issue without -sameq
[19:54] <pasteeater> can you upload foo.mp4 somewhere?
[19:54] <teratorn> sure, one sec
[19:55] <teratorn> http://teratorn.org/code/foo.mp4
[19:58] <teratorn> pasteeater: foo.mp4 might be really screwy - I dunno
[20:01] <pasteeater> it has some weird frame rate issues
[20:03] <pasteeater> you can add -r 26.28 and it should encode
[20:03] <pasteeater> what did foo.mp4 come from?
[20:04] <teratorn> a rather complex capturing/compositing/encoding pipeline
[20:04] <teratorn> which, apparently, has problems
[20:07] <teratorn> pasteeater: thanks, that helps. do you know what specifically is wrong with foo.mp4?
[20:27] <Mista_D> Can I utilize all 16 cores to encode a 2 pass 720p file from 1080p source? `-threads 0` only uses 200% instead of 1600% according to top.
[20:31] <Mavrik> Mista_D, for which pass_
[20:31] <Mavrik> pass 1 will probably use only 1 or 2 cores
[20:31] <Mavrik> pass 2 should use at least half
[20:36] <JackD> pasteeater:
[20:36] <JackD> can you give me a hand ?
[20:40] <cbsrobot> JackD: I guess your card runs at a certain speed
[20:40] <cbsrobot> eh framerate
[20:40] <cbsrobot> why not do it with a for loop ?
[20:41] <cbsrobot> or while
[20:41] <JackD> but it could be slow, no?
[20:42] <cbsrobot> well you want it to grab one fram per second - no?
[20:43] <JackD> yes, but have no way to make ffmpeg drop some frames?
[20:44] <cbsrobot> i=0; for ((;;)); do i=`expr $i + 1`; ffmpeg -f vfwcap -s 640x480 -vrames 1 -i 0 output-$i.jpg; done;
[20:44] <cbsrobot> untested
[20:45] <cbsrobot> ah shit - add: sleep 1;
[20:46] <JackD> ok ok ok you convinced me, I will use a script or compile a c source,
[20:47] <cbsrobot> I never used vfwcap btw
[20:47] <JackD> I see
[20:47] <JackD> I hnever use too
[20:47] <JackD> thankyou 
[20:48] <cbsrobot> or maybe
[20:48] <cbsrobot> use filters
[20:48] <cbsrobot> like -vf select='not(mod(n\,15))'
[20:48] <cbsrobot> for 15fps input
[20:49] <cbsrobot> fmpeg -f vfwcap -s 640x480 -vrames 1 -i 0 -vf select='not(mod(n\,15))' output%d.jpg
[20:49] Action: cbsrobot copied your command line ...
[20:49] <cbsrobot> and drop the vframes - lol
[20:50] <cbsrobot> I should stop writing wrong stuff
[20:57] <JackD> cbsrobot: I tried the filtes but with no much sucess, the for ideia is not too hungry then I will chose it by now. Thanks 
[20:58] <cbsrobot> JackD: I guess the filters version is saner
[20:59] <cbsrobot> if you use this, do you get an output ?
[20:59] <cbsrobot> ffmpeg -f vfwcap -s 640x480 -i 0 test.mpg
[20:59] <JackD> yes
[21:00] <cbsrobot> whats the framerate of the output file ?
[21:00] <JackD> - 0 is a surveillance camera with 30fps
[21:01] <JackD> -i 0 *
[21:01] <JackD> 29fps
[21:01] <cbsrobot> ffmpeg -f vfwcap -s 640x480 -i 0 -vf select='not(mod(n\,29))' output%d.jpg
[21:01] <cbsrobot> do you get jpeg files when running ^
[21:02] Action: cbsrobot thinks you can even use output-%05d.jpg
[21:02] <JackD> I think I will get error with this, trying...
[21:03] <JackD> too many arguments in mod
[21:03] <cbsrobot> ah
[21:04] <cbsrobot> ffmpeg -f vfwcap -s 640x480 -i 0 -vf "select=not(mod(n\,29))" output%d.jpg
[21:05] <JackD> 217 jpg files in 10s
[21:06] <cbsrobot> try
[21:07] <cbsrobot> hmm
[21:08] <cbsrobot> still thinking
[21:08] <JackD> I still trying some idiot things here
[21:08] <cbsrobot> ah
[21:09] <cbsrobot> ffmpeg -f vfwcap -r 30 -s 640x480 -i 0 -vf "select=not(mod(n\,29))" output%d.jpg
[21:10] <JackD> 217 files for 10s
[21:10] <cbsrobot> shit
[21:10] <JackD> a lot of
[21:10] <cbsrobot> ffmpeg -f vfwcap -r 30 -s 640x480 -i 0 -vf "select=not(mod(n\,100))" output%d.jpg
[21:11] <cbsrobot> do you get less jpegs ?
[21:12] <JackD> i tried with 300 and i get 2 jpgs in 10 seconds
[21:12] <JackD> what n\ means?
[21:12] <cbsrobot> n = input frame number
[21:12] <cbsrobot> \ = escapecharacter for ,
[21:13] <cbsrobot> , is used to separate filters aswll
[21:13] <JackD> not is the bitewise NOt right?
[21:13] <cbsrobot> not is just not
[21:13] <JackD> ahh ahahahahaha
[21:14] <cbsrobot> mod(n,100) is only 0, if you hit the 10th frame
[21:14] <cbsrobot> now inverse that
[21:14] <cbsrobot> you can quote me on that !
[21:14] <cbsrobot> lol
[21:14] <relaxed> JackD: for highest quality jpgs add -qscale 1
[21:14] <cbsrobot> *100th
[21:16] <JackD> inverse?
[21:16] <JackD> aaaaa
[21:17] <JackD> relaxed: if I use -qscale 6 i get less quality right?
[21:17] <cbsrobot> JackD: yes
[21:18] <relaxed> JackD: correct, I think the scale is 1 - 31 (or is it 51)?
[21:18] <relaxed> it's 31
[21:18] <cbsrobot> imho 31
[21:18] <cbsrobot> or 42 ?
[21:19] <cbsrobot> it's 42
[21:19] <relaxed> that's the answer to everything! :)
[21:19] <cbsrobot> I heard it on TV !
[21:23] <JackD> if I use 100 or above I get a frame every 5s
[21:23] <JackD> bellow it the result is the same 217 files
[21:26] <cbsrobot> JackD: pastebin the lines for ffmpeg -f vfwcap -r 30 -s 640x480 -i 0 -t 10 -vf showinfo -f null -y /dev/null
[21:36] <JackD> what info do you want from it?
[21:36] <JackD> it repeated the same line 
[21:38] <cbsrobot> all to pastebin.com
[21:38] <JackD> http://codepad.org/bU7Bq4xW
[21:42] <JackD> have ffmpeg a silent option?
[21:42] <cbsrobot> aaaah codepad wraps the line ... that sucks
[21:43] <cbsrobot> JackD: pastebin please
[21:43] Action: cbsrobot too lazy
[21:43] <JackD> no no it was DOS problem
[21:43] <JackD> not codepad
[21:44] <JackD> I will try to enlarge
[21:44] <cbsrobot> ah
[21:44] <cbsrobot> np
[21:50] <JackD> http://srb.no-ip.org:9073/pastebin/ffmpeg.html ?
[21:51] <cbsrobot> JackD: cant connect
[21:53] <JackD> I will ask someone to put it on pastebin ...
[21:53] <cbsrobot> JackD: np
[21:53] <cbsrobot> JackD: nevermind
[21:54] <cbsrobot> I see the value of n
[21:54] <cbsrobot> actually it should work as is
[21:55] <cbsrobot> try to tune the mod value to your needs
[21:55] <cbsrobot> ffmpeg -f vfwcap -r 30 -s 640x480 -i 0 -vf "select=not(mod(n\,50))" output-%05d.jpg
[21:57] <JackD> anyway http://pastebin.com/CUmTS6Za
[21:58] <cbsrobot> JackD: also wrapped
[21:59] <cbsrobot> nevermind
[21:59] <JackD> as I said is is because of the DOS not because of the pastebin or codepad
[21:59] <JackD> I get 8 or 80
[22:00] <JackD> I can not tune it or it generates 217 or 1frame
[22:04] <cbsrobot> JackD: lol
[22:28] <davidOmatic> hi! can ffmpeg extract tracks from audio cds?
[22:50] <osearth> "Dying is winning, the longer you embrace life and joy the more you won." - ose
[23:01] <Mista_D> What is the basic difference between slice and frame thread modes? Does/can they affect quality of encoding at all?
[23:04] <cbsrobot> Mista_D: affect quality - hopefully not
[23:04] <cbsrobot> slice is like a frame split in to smaller chunks
[23:05] <cbsrobot> frame threading is just one frame per thread
[23:07] <Mista_D> cbsrobot: thanks. I thought when only partial frame inforamtion is available to the encoder it can not really achieve good compression as its can't reuse the parts of video that are the same...
[23:11] <Mista_D> Is there a way to extract stdout & errout with multiple outputs?? eg: `./avconv -i 1080.ts -s 1280x720 720p.mp4 2>&1 >720.log -s 320x176 -r 12 preview.mp4  2>&1 >preview.log` -- doesn't work, logs are the same.
[23:16] <cbsrobot> use ....mp4 > preview.log 2>&1 error.log
[23:16] <cbsrobot> ah - multiple ... no idea
[23:22] <_stev3> hey - can anyone tell me how i can get ffmpeg to return the resolution of a file?
[23:23] <pasteeater> _stev3: ffmpeg -i input 2>&1 | awk '/Video/{print $7}'
[23:25] <_stev3> that returns '[PAR'
[23:26] <pasteeater> probably depends on your format
[23:26] <cbsrobot> hehe
[23:27] <pasteeater> or maybe ffmpeg -i input 2>&1 | grep -i video | grep -o "[0-9]*x[0-9]*"
[23:27] <pasteeater> not very efficient
[23:28] <_stev3> nice, thanks :D
[23:28] <cbsrobot> ffprobe -show_streams input | egrep "width|height"
[00:00] --- Sat Dec  3 2011